Initiating A Collective Operation In A Parallel Computer

1. A method of initiating a collective operation in a parallel computer, the parallel computer comprising a plurality of compute nodes, the compute nodes coupled for data communications by one or more data communications networks, the compute nodes organized in an operational group for collective operations with one compute node assigned as a root node of the operational group, the method comprising: identifying, by a non-root compute node, a collective operation to execute in the operational group of compute nodes; initiating, by the non-root compute node, execution of the collective operation amongst the compute nodes of the operational group including: sending, by the non-root compute node to one or more of the other compute nodes in the operational group, an active message, the active message comprising information configured to initiate execution of the collective operation amongst the compute nodes of the operational group; and executing, by the compute nodes of the operational group, the collective operation.

2. The method of claim 1 , wherein: the compute nodes of the operational group are configured in a tree topology; initiating execution of the collective operation further comprises initiating execution of a broadcast operation; and sending the active message to one or more of the other compute nodes further comprises sending by the non-root compute node to the root compute node an active message instructing the root node to send, to each of the root node's children, contribution data and another instruction for the root node's children to forward the same contribution data and the same instruction to the children of the root node's children.

3. The method of claim 2 , wherein the contribution data represents a notification to cease sending error messages.

4. The method of claim 1 , wherein: the compute nodes of the operational group are configured in a tree topology; initiating execution of the collective operation further comprises initiating execution of a reduce operation; and sending the active message to one or more of the other compute nodes further comprises sending, by the non-root compute node to each of the leaf nodes positioned at the bottom of the tree topology, an active message instructing each of the leaf nodes to forward data to the leaf node's parent, perform an operation on the data, and forward an instruction to the leaf node's parent to carry out the same steps.

5. The method of claim 4 , wherein the reduction operation effects event consolidation in the operational group of compute nodes.

6. The method of claim 4 , wherein the reduction operation effects event suppression in the operational group of compute nodes.

7. An apparatus for initiating a collective operation in a parallel computer, the parallel computer comprising a plurality of compute nodes, the compute nodes coupled for data communications by one or more data communications networks, the compute nodes organized in an operational group for collective operations with one compute node assigned as a root node of the operational group, the apparatus comprising a computer processor, a computer memory operatively coupled to the computer processor, the computer memory having disposed within it computer program instructions that, when executed by the computer processor, cause the apparatus to carry out the steps of: identifying, by a non-root compute node, a collective operation to execute in the operational group of compute nodes; initiating, by the non-root compute node, execution of the collective operation amongst the compute nodes of the operational group including: sending, by the non-root compute node to one or more of the other compute nodes in the operational group, an active message, the active message comprising information configured to initiate execution of the collective operation amongst the compute nodes of the operational group; and executing, by the compute nodes of the operational group, the collective operation.

8. The apparatus of claim 7 , wherein: the compute nodes of the operational group are configured in a tree topology; initiating execution of the collective operation further comprises initiating execution of a broadcast operation; and sending the active message to one or more of the other compute nodes further comprises sending by the non-root compute node to the root compute node an active message instructing the root node to send, to each of the root node's children, contribution data and another instruction for the root node's children to forward the same contribution data and the same instruction to the children of the root node's children.

9. The apparatus of claim 8 , wherein the contribution data represents a notification to cease sending error messages.

10. The apparatus of claim 7 , wherein: the compute nodes of the operational group are configured in a tree topology; initiating execution of the collective operation further comprises initiating execution of a reduce operation; and sending the active message to one or more of the other compute nodes further comprises sending, by the non-root compute node to each of the leaf nodes positioned at the bottom of the tree topology, an active message instructing each of the leaf nodes to forward data to the leaf node's parent, perform an operation on the data, and forward an instruction to the leaf node's parent to carry out the same steps.

11. The apparatus of claim 10 , wherein the reduction operation effects event consolidation in the operational group of compute nodes.

12. The apparatus of claim 10 , wherein the reduction operation effects event suppression in the operational group of compute nodes.

13. A computer program product for initiating a collective operation in a parallel computer, the parallel computer comprising a plurality of compute nodes, the compute nodes coupled for data communications by one or more data communications networks, the compute nodes organized in an operational group for collective operations with one compute node assigned as a root node of the operational group, the computer program product disposed upon a computer readable medium, the computer program product is not a signal, the computer program product comprising computer program instructions that, when executed, cause a computer to carry out the steps of: identifying, by a non-root compute node, a collective operation to execute in the operational group of compute nodes; initiating, by the non-root compute node, execution of the collective operation amongst the compute nodes of the operational group including: sending, by the non-root compute node to one or more of the other compute nodes in the operational group, an active message, the active message comprising information configured to initiate execution of the collective operation amongst the compute nodes of the operational group; and executing, by the compute nodes of the operational group, the collective operation.

14. The computer program product of claim 13 , wherein: the compute nodes of the operational group are configured in a tree topology; initiating execution of the collective operation further comprises initiating execution of a broadcast operation; and sending the active message to one or more of the other compute nodes further comprises sending by the non-root compute node to the root compute node an active message instructing the root node to send, to each of the root node's children, contribution data and another instruction for the root node's children to forward the same contribution data and the same instruction to the children of the root node's children.

15. The computer program product of claim 14 , wherein the contribution data represents a notification to cease sending error messages.

16. The computer program product of claim 13 , wherein: the compute nodes of the operational group are configured in a tree topology; initiating execution of the collective operation further comprises initiating execution of a reduce operation; and sending the active message to one or more of the other compute nodes further comprises sending, by the non-root compute node to each of the leaf nodes positioned at the bottom of the tree topology, an active message instructing each of the leaf nodes to forward data to the leaf node's parent, perform an operation on the data, and forward an instruction to the leaf node's parent to carry out the same steps.

17. The computer program product of claim 16 , wherein the reduction operation effects event consolidation in the operational group of compute nodes.

18. The computer program product of claim 16 , wherein the reduction operation effects event suppression in the operational group of compute nodes.

19. The computer program product of claim 13 wherein the computer readable medium comprises a storage medium.